How kartoza/docker-geoserver works
Project type and purpose
• This is a Docker‐based deployment project (“Kartoza docker-geoserver”) that packages GeoServer (a Java/Tomcat GIS server) with configurable extensions and optional clustering using PostGIS and ActiveMQ.
• Its purpose is to provide a production-ready GeoServer container, backed by PostGIS, with environment variables and mountable configs to customize behavior, enable extensions, clustering, SSL, reverse proxy, backups, etc.
File structure
• Top-level Docker assets:
• Dockerfile, docker-compose.yml variants, .env, .dockerignore
• build_data/ – default GeoServer and Tomcat configs, plugins lists, SQL scripts, XML templates
• scripts/ – container entrypoint, setup and helper shell scripts
• clustering/ – example Dockerfiles and compose files for JMS/ActiveMQ clustering
• sites-enabled/nginx.conf – sample Nginx reverse-proxy config
• CI/CD:
• .github/workflows/ – build and deploy-image pipelines
• Scenario tests:
• scenario_tests/ – end-to-end tests using docker-compose for various features (backup, clustering, disk quota, JNDI, etc.)
• Documentation:
• docs/ – static HTML docs for usage, environment variables, Kubernetes, mounting configs
• Resources & overlays for custom file overlays
README for additional insights
• Describes main runtime components: GeoServer on Tomcat, PostGIS backend, optional Nginx or Kubernetes front ends, JMS clustering plugin with ActiveMQ broker.
• Describes how environment variables drive plugin activation, disk quota, SSL, JNDI, proxy settings, context root.
• Mentions Helm charts for Kubernetes, CI scanning with Trivy, upgrade scripts.
Elements
• Clear labels: container names (geoserver, db, broker), protocol (HTTP, JDBC, JMS)
• Containers (rectangles in blue)
Detail
• Config files in build_data vs override mount points
• How environment variables (from .env) map to container startup behavior
• Scenario tests as a separate component that spins up compose stacks and connects to the services
• Don’t overcrowd: group similar items (all .yml files under clustering as “clustering examples”)
When drawing, start with a high‐level overview (User → Proxy → GeoServer → DB) then drill into container internals (Tomcat, build_data, scripts, plugins, env vars) as needed. Use layering: Network layer, Container layer, File/Volume layer, CI/CD layer.